Honglin Ye wrote:
I think search over multiple collections of similar document is a needed
feature.
Well, it was and still is on Xindice TODO list :-)
By partitioning the database into smaller collections may improve the performance.
I'm curious, why do you think so? Is there any reason or observation for this?
Please bear in mind also that lots of collections will exhaust your system resources, particularly, file handles (see archives - somebody already hit that limit, I guess he had thousands of collections).
...
I modified XPathQuery.java to do multiple collection search.
...
I am new to xindice and am not sure if this modification make sense or if it has impact on other modules.
Thanks for your contribution; I'll review your changes as soon as I get some time.
Vadim
